Coreweave outage history

August 20, 2026 5:12PM UTC Investigating - We are investigating a loss of production GPU capacity affecting a subset (19) of GB300 racks in DH204. Affected nodes have intermittently gone unhealthy and been evicted; a number have been taken out of production for recovery. Affected: GB300 compute Current impact: Several racks are below the 16-node threshold for a full NVL64 domain, so multi-node jobs targeting those racks may fail to place or may be interrupted. Capacity in other datahalls and regions is unaffected. The first node losses were observed at approximately 2026-08-19 19:00 UTC. Fleet Operations has been recovering nodes by reboot since then, and additional engineering teams were engaged at 16:30 UTC today. We will keep you posted on further details as we progress. Thank you for your patience and cooperation while we are investigating.

August 13, 2026 11:21PM UTC Investigating - Beginning at 22:51 UTC on 13 August 2026, we identified a network connectivity issue affecting internal service communication within the EU-SOUTH-03B availability zone. Customers in this availability zone may experience intermittent errors or delays when performing cluster control-plane operations, such as Kubernetes API requests. Running workloads and compute nodes are not currently showing impact. Our networking teams are actively engaged and investigating. Other availability zones and regions are unaffected. August 14, 2026 12:56AM UTC Resolved - Between 22:51 UTC on 13 August and 00:38 UTC on 14 August 2026, customers with resources in the EU-SOUTH-03B availability zone may have experienced intermittent errors performing cluster control-plane operations, degraded cluster access, and brief periods where nodes reported as unavailable. The cause was a hardware fault affecting network links within the availability zone. Our network engineering team removed the affected links from service, and connectivity has been stable since 00:38 UTC. Node capacity was not lost during this event, and no other availability zones or regions were affected. This incident is now resolved.

August 10, 2026 11:00PM UTC Investigating - Hello, We are actively investigating a storage issue that may be impacting our storage services in US-EAST-02A. As the situation evolves, we will continue to update our customers on the matter. If you have any questions, we encourage you to reach out to support by email: [email protected] We apologize for any inconvenience this may cause and thank you for your patience while we work to resolve the matter in a timely manner. Thank you, CoreWeave Support August 10, 2026 11:28PM UTC Identified - We have identified the root cause of the issue and are actively working on a fix. We will update you when we have pushed a fix. August 11, 2026 7:03PM UTC Resolved - The issue is resolved, and our services are now operational. If you are continuing to experience issues, please reach out to [email protected].

August 9, 2026 3:31PM UTC Investigating - We are currently investigating an issue related to node delivery/reboot failures occurring in the RNO2 region. As the situation evolves, we will continue to provide our customers with updates on the matter. If you have any questions, we encourage you to reach out to support by email: [email protected] We apologize for any inconvenience this may cause and thank you for your patience while we work to resolve the matter in a timely manner. Thank you, CoreWeave Support August 9, 2026 5:22PM UTC Identified - The root cause has been identified and we are working on a resolution at this time. August 9, 2026 5:41PM UTC Monitoring - This incident has been mitigated. Node delivery and node reboot operations have now returned to normal. If you continue to experience any issues, please reach out to support: [email protected]. We apologize for the disruption this caused and thank you for your patience while we worked to resolve the matter. Thank you, CoreWeave Support August 9, 2026 11:05PM UTC Resolved - This incident is now resolved and our services are back operational. If you have any further questions please reach out to [email protected].

August 3, 2026 7:59PM UTC Investigating - We are investigating a service degradation in the US-EAST-04A region that may result in disruptions to cluster access/connectivity. As the situation evolves, we will continue to provide updates. If you have any questions, please contact [email protected]. We apologize for any inconvenience this may cause and appreciate your patience while we work to resolve the issue as quickly as possible. August 4, 2026 4:53PM UTC Identified - We have identified a power issue which has brought a swath of nodes offline. We are working to bring spare capacity back to production. August 6, 2026 8:36PM UTC Resolved - We believe the issue is resolved, and our services are now operational. If you are continuing to experience issues, please reach out to [email protected].

July 29, 2026 7:03PM UTC Scheduled - CoreWeave will be performing planned electrical maintenance at our US-EAST-04A facility from Monday, August 3, 2026, at 12:00 PM UTC through Thursday, August 6, 2026, at 7:00 PM UTC. This work upgrades the permanent power infrastructure for the DH3 B3 electrical lineup at CSZ01 (CTP-1). Phase 1 (August 3 through August 4): Utility feeder cutover and UPS testing. No planned impact to customer workloads. Electrical redundancy on the B3 lineup will be temporarily reduced during this phase. Phase 2 (August 4, approximately 5:00 PM UTC / 1:00 PM ET, through August 6, approximately 5:00 PM UTC / 1:00 PM ET): B3 PDU shutdown and permanent power cutover. A subset of compute nodes in US-EAST-04A on the DH3 B3 lineup will experience hard power loss for approximately 48 hours. CKS nodes will go NotReady; bare-metal nodes will power off. Network infrastructure in the region is not expected to lose power but may operate at reduced electrical redundancy. The customers affected in Phase 2 will be contacted directly prior to the maintenance. Impact: During Phase 2, affected compute capacity in US-EAST-04A will be unavailable for the duration of the PDU shutdown window. We do not expect impact to the Kubernetes control plane or to capacity outside the affected DH3 B3 node set. Other infrastructure in US-EAST-04A is unaffected. We will be actively monitoring the situation during this time. We appreciate your patience and understanding. If you have any further questions or require assistance, please don't hesitate to reach out to our support team at [email protected]. August 4, 2026 4:17PM UTC Active - Maintenance is in progress August 5, 2026 4:14PM UTC Active - This maintenance work has started August 5, 2026 8:07PM UTC Completed - Maintenance complete. Facilities work on the US-EAST-04A DH3 B3 electrical lineup at CSZ01 (CTP-1) has finished, and affected compute capacity has been restored. Power has been returned to the B3 PDUs and racks. We appreciate your patience during this maintenance window. If you continue to experience any issues, please contact [email protected].

August 4, 2026 10:56PM UTC Investigating - We are investigating a networking hardware issue in US-EAST-04A affecting connectivity for a subset of nodes. Out of an abundance of caution, impacted nodes have been temporarily removed from service while our engineers work to restore connectivity. Customers may see a small number of nodes become unavailable or unschedulable during this time. August 4, 2026 11:53PM UTC Identified - Our engineers have identified the issue and remediation is underway. We will provide another update within the next hour or sooner. August 5, 2026 12:25AM UTC Monitoring - The issue has been resolved and affected nodes have returned to service. We will continue to monitor. August 5, 2026 12:27AM UTC Resolved - We believe the issue is resolved, and our services are now operational. If you are continuing to experience issues, please reach out to [email protected].

August 2, 2026 1:08AM UTC Identified - We have identified the root cause of the issue and are actively working on a fix. We will update you when we have pushed a fix. August 2, 2026 1:23AM UTC Identified - We have identified the cause of an issue affecting Object Storage access key creation globally. Customers using SAML or OIDC-based access key creation may currently be unable to create new access keys. Our engineering team is actively working on mitigation and evaluating additional protective measures to restore normal service. We will continue to provide updates as progress is made. August 2, 2026 3:33AM UTC Resolved - CoreWeave Object Storage access key creation has been restored and is now ready and in working order. We appreciate your patience.

July 31, 2026 3:52PM UTC Identified - Hello, We are actively investigating elevated error rates affecting requests to some models available through CoreWeave Serverless and W&B Inference. Customers may experience failed requests. As the situation evolves, we will continue to provide our customers with updates on the matter. If you have any questions, we encourage you to reach out to support by email: [email protected] We apologize for the inconvenience and appreciate your patience. Thank You, CoreWeave Support July 31, 2026 5:01PM UTC Monitoring - Hello, The issue affecting some models available through CoreWeave Serverless and W&B Inference has been mitigated. Service has recovered, and we are monitoring performance to ensure continued stability. We will provide another update when monitoring is complete. We apologize for the inconvenience and appreciate your patience. Thank You, CoreWeave Support August 1, 2026 12:37AM UTC Resolved - Hello, The issue causing elevated error rates for some models available through CoreWeave Serverless and W&B Inference has been resolved. All affected models are now responding successfully. If you continue to experience issues, please contact [email protected]. We apologize for the inconvenience and appreciate your patience. Thank you, CoreWeave Support

July 29, 2026 1:57AM UTC Investigating - We are currently investigating an elevated rate of failed requests to CoreWeave AI Object Storage in the US-EAST-06 region. July 29, 2026 1:58AM UTC Identified - We have identified the issue and are continuing to work on a resolution at this time. July 29, 2026 5:34AM UTC Monitoring - We believe the issue has been mitigated, and we are continuing to monitor the affected services. If you are experiencing an issue or have any further questions please reach out to [email protected]. July 29, 2026 6:47AM UTC Resolved - We believe the issue is resolved, and our services are now operational. If you are continuing to experience issues, please reach out to [email protected].

July 21, 2026 10:03PM UTC Investigating - We're currently investigating an issue that is preventing metrics and dashboards from loading for resources in our US-EAST regions. Customers may be unable to view cluster and node metrics in Grafana, and monitoring queries for US-EAST resources are returning errors. This is limited to metrics visibility only - It does not affect the health, availability, or performance of your running workloads, clusters, or nodes - compute continues to operate normally. Other regions (US-WEST and US-CENTRAL) are unaffected and their dashboards remain fully available. Our engineers are actively investigating and working to restore metrics access as quickly as possible. We apologize for the disruption and appreciate your patience. July 21, 2026 10:38PM UTC Monitoring - Metrics and dashboards for our US-EAST region have been restored, and monitoring queries are returning results normally again. Our engineers have applied a fix and are continuing to monitor closely to confirm full stability. As noted earlier, this issue was limited to metrics visibility and did not affect the health, availability, or performance of running workloads, clusters, or nodes. If you're still seeing gaps or errors when loading US-EAST metrics, please let us know. We'll post a final update once we've confirmed the region is fully stable. Thank you for your patience during the disruption. July 21, 2026 10:50PM UTC Resolved - We believe the issue is resolved, and our services are now operational. If you are continuing to experience issues, please reach out to [email protected].

July 21, 2026 8:30PM UTC Investigating - Hello , We're investigating an issue affecting metrics and logs in EU-NORTH. Customers with workloads in EU-NORTH-04 and EU-NORTH-05 may see missing metrics and logs. Compute and running workloads are not affected. Our team is actively working on to remediate the issue. We will keep you posted as we progress. We apologise for any inconvenienve and thank you for your patience! Thanks, CoreWeave Support July 21, 2026 8:41PM UTC Monitoring - A fix has been applied and connectivity for the EU-NORTH region has been restored. Metrics and logs for workloads in EU-NORTH-04 and EU-NORTH-05 are catching up, and we are monitoring to confirm full recovery. Compute and running workloads were not affected. We will confirm once the issue is fully resolved. July 21, 2026 9:02PM UTC Resolved - This issue is resolved. For approximately one hour, metrics and logs for workloads in EU-NORTH-04 and EU-NORTH-05 were degraded due to an upstream network transport issue affecting the path to our observability backend. As of 20:30 UTC, connectivity has been restored, metrics and logs have caught up, and services are operating normally. Compute and running workloads were not affected at any point. We apologize for the inconvenience and thank you for your patience.

July 15, 2026 1:07PM UTC Scheduled - Details: CoreWeave will be performing maintenance on Thursday, July 16, 2026 from 16:00 UTC to 03:00 UTC, Saturday, July 18, 2026 that will affect our VAST storage cluster. Impact: No downtime is expected. During the maintenance window you may experience brief periods of degraded storage performance or elevated latency while drives are upgraded. We will be actively monitoring the situation during this time. We appreciate your patience and understanding. If you have any further questions or require assistance, please don't hesitate to reach out to our support team at [email protected] This activity is deemed Planned Maintenance under the CoreWeave Terms of Service Maintenance Policy ( https://docs.coreweave.com/policies/terms-of-service/maintenance-policy ) July 16, 2026 4:00PM UTC Active - Scheduled maintenance is starting. July 18, 2026 3:00AM UTC Completed - Scheduled maintenance is complete.

July 15, 2026 1:09PM UTC Scheduled - Details: CoreWeave will be performing maintenance on Thursday, July 16, 2026 from 16:00 UTC to 03:00 UTC, Saturday, July 18, 2026 that will affect our VAST storage cluster. Impact: No downtime is expected. During the maintenance window you may experience brief periods of degraded storage performance or elevated latency while drives are upgraded. We will be actively monitoring the situation during this time. We appreciate your patience and understanding. If you have any further questions or require assistance, please don't hesitate to reach out to our support team at [email protected] This activity is deemed Planned Maintenance under the CoreWeave Terms of Service Maintenance Policy ( https://docs.coreweave.com/policies/terms-of-service/maintenance-policy ) July 16, 2026 4:00PM UTC Active - Scheduled maintenance is starting. July 18, 2026 3:00AM UTC Completed - Scheduled maintenance is complete.

July 15, 2026 4:04PM UTC Investigating - Hello, We are actively investigating an issue impacting CW API Services globally. As the situation evolves, we will continue to provide our customers with updates on the matter. If you have any questions, we encourage you to reach out to support by email: [email protected] We apologize for any inconvenience this may cause and thank you for your patience while we work to resolve the matter in a timely manner. July 15, 2026 4:50PM UTC Identified - We have identified the issue as a disruption to CW managed authentication. While actual health of cluster(s) are fine, anyone using CW managed authentication will be unable to authenticate due to an outage in our authentication service. While the authentication service is down, if you are able to bypass your CW managed Authentication, you will still be able to access your cluster(s). We are urgently working on a fix to our authentication service in parallel and will continue to post updates to this status page. July 15, 2026 5:52PM UTC Monitoring - We believe this incident has been mitigated. A fix went out to our authentication service and we're seeing a return to cluster/console access . We are continuing to monitor to ensure stability. If you continue to experience any issues, please reach out to support: [email protected] July 16, 2026 10:44AM UTC Resolved - Hello, The service disruption affecting CW managed authentication is now resolved and our systems are operating normally. If you continue to experience problems or notice any residual impact, please contact [email protected] so we can investigate further. Thank you for your patience while we worked to restore service. We apologise for any inconvenience caused. Thank you, CoreWeave Support Team

July 14, 2026 3:48PM UTC Scheduled - Details: CoreWeave will be performing maintenance on Wednesday, July 15, 2026 from 16:00 UTC to 20:00 UTC that will affect our VAST storage cluster. Impact: No downtime is expected. During the maintenance window you may experience brief periods of degraded storage performance or elevated latency while drives are upgraded. We will be actively monitoring the situation during this time. We appreciate your patience and understanding. If you have any further questions or require assistance, please don't hesitate to reach out to our support team at [email protected] This activity is deemed Planned Maintenance under the CoreWeave Terms of Service Maintenance Policy ( https://docs.coreweave.com/policies/terms-of-service/maintenance-policy ) July 15, 2026 4:00PM UTC Active - Scheduled maintenance is starting. July 16, 2026 4:00AM UTC Completed - Scheduled maintenance is complete.

July 14, 2026 3:46PM UTC Scheduled - Details: CoreWeave will be performing maintenance on Wednesday, July 15, 2026 from 16:00 UTC to 20:00 UTC that will affect our VAST storage cluster. Impact: No downtime is expected. During the maintenance window you may experience brief periods of degraded storage performance or elevated latency while drives are upgraded. We will be actively monitoring the situation during this time. We appreciate your patience and understanding. If you have any further questions or require assistance, please don't hesitate to reach out to our support team at [email protected] This activity is deemed Planned Maintenance under the CoreWeave Terms of Service Maintenance Policy ( https://docs.coreweave.com/policies/terms-of-service/maintenance-policy ) July 15, 2026 4:00PM UTC Active - Scheduled maintenance is starting. July 16, 2026 4:00AM UTC Completed - Scheduled maintenance is complete.

July 14, 2026 3:45PM UTC Scheduled - Details: CoreWeave will be performing maintenance on Wednesday, July 15, 2026 from 16:00 UTC to 20:00 UTC that will affect our VAST storage cluster. Impact: No downtime is expected. During the maintenance window you may experience brief periods of degraded storage performance or elevated latency while drives are upgraded. We will be actively monitoring the situation during this time. We appreciate your patience and understanding. If you have any further questions or require assistance, please don't hesitate to reach out to our support team at [email protected] This activity is deemed Planned Maintenance under the CoreWeave Terms of Service Maintenance Policy ( https://docs.coreweave.com/policies/terms-of-service/maintenance-policy ) July 15, 2026 4:00PM UTC Active - Scheduled maintenance is starting. July 16, 2026 4:00AM UTC Completed - Scheduled maintenance is complete.

July 13, 2026 9:31PM UTC Scheduled - Details: CoreWeave will be performing maintenance on Tuesday, July 14, 2026 from 22:00 UTC to 02:00 UTC (July 15) that will affect our VAST storage cluster. Impact: No downtime is expected. During the maintenance window you may experience brief periods of degraded storage performance or elevated latency while drives are upgraded. We will be actively monitoring the situation during this time. We appreciate your patience and understanding. If you have any further questions or require assistance, please don't hesitate to reach out to our support team at [email protected] This activity is deemed Planned Maintenance under the CoreWeave Terms of Service Maintenance Policy ( https://docs.coreweave.com/policies/terms-of-service/maintenance-policy ) July 14, 2026 10:00PM UTC Active - Scheduled maintenance is starting. July 15, 2026 2:00AM UTC Completed - Scheduled maintenance is complete.

July 13, 2026 9:34PM UTC Scheduled - Details: CoreWeave will be performing maintenance on Tuesday, July 14, 2026 from 22:00 UTC to 02:00 UTC (July 15) that will affect our VAST storage cluster. Impact: No downtime is expected. During the maintenance window you may experience brief periods of degraded storage performance or elevated latency while drives are upgraded. We will be actively monitoring the situation during this time. We appreciate your patience and understanding. If you have any further questions or require assistance, please don't hesitate to reach out to our support team at [email protected] This activity is deemed Planned Maintenance under the CoreWeave Terms of Service Maintenance Policy ( https://docs.coreweave.com/policies/terms-of-service/maintenance-policy ) July 14, 2026 10:00PM UTC Active - Scheduled maintenance is starting. July 15, 2026 2:00AM UTC Completed - Scheduled maintenance is complete.